From Data to Hope: Deep Neural Network-Based Prediction of Poisoning (DNNPPS) Suicide Cases.

Abstract

Background: Suicide is a critical global issue with profound social and economic consequences. Implementing effective prevention strategies is essential to alleviate these impacts. Deep neural network (DNN) algorithms have gained significant traction in health sectors for their predictive capability. We looked at the potential of DNNs to predict suicide cases.

Methods: A descriptive-analytical, cross-sectional study was conducted to analyze suicide data using a deep neural network predictive prevention system (DNNPPS). The analysis utilized a suicide dataset comprising 1,500 data points, provided by a health research center in Kerman, Iran, spanning the years 2019-2022.

Results: Factors such as history of psychiatric hospitals, days of the week, and job were identified as the most important risk factors for predicting suicide attempts. Promising results were obtained by applying the DNNPPS model to a dataset of 1453 individuals with a history of suicide. The problem was approached as a binary classification task, with suicide history as the target variable. We performed preprocessing techniques, including class balancing, and constructed a DNN model using a sequential architecture with four dense layers.

Conclusion: The success of the DNN algorithm depends on the quality and quantity of data, as well as the model's architecture. High-quality data should be accurate, representative, and relevant, while a large dataset enables the DNN to learn more features. In our study, the DNNPPS model performed well, achieving an F1-score of 91%, which indicates high accuracy in predicting suicide cases and a good balance between precision and recall.